FOLIA PARASITOLOGICA 57[3]: 223–231, 2010 © Institute of Parasitology, Biology Centre ASCR ISSN 0015-5683 (print), ISSN 1803-6465 (online) http://www.paru.cas.cz/folia/

New evidence on a cold case: trophic transmission, distribution and host-specificity in Hedruris spinigera (Nematoda: Hedruridae)

José L. Luque1, Fabiano M. Vieira2, Kristin Herrmann3, Tania M. King3, Robert Poulin3 and Clément Lagrue4

1 Departamento de Parasitologia , Universidade Federal Rural do Rio de Janeiro, Caixa Postal 74.508, CEP 23851-970 Seropédica, RJ, Brasil; 2 Curso de Pós-Graduação em Biologia Animal, Universidade Federal Rural do Rio de Janeiro, Seropédica, RJ, Brasil; 3 Department of Zoology, University of Otago, P.O. Box 56, Dunedin 9054, New Zealand; 4 Laboratoire d’Ecologie Fonctionnelle (EcoLab, CNRS UMR 5245), 29 Rue Jeanne Marvig, BP 24349, 31055 Toulouse cedex 4, France

Abstract: The life cycle of Hedruris spinigera Baylis, 1931 (Nematoda: Hedruridae) is determined here with the first formal iden- tification of the parasite’s intermediate host: the amphipod excavatum Thomson. Adult H. spinigera are redescribed from specimens collected from the stomach of fishes, Retropinna retropinna (Richardson) and Aldrichetta forsteri (Va- lenciennes), from Lake Waihola, New Zealand. Immature adults of the parasite collected from intermediate hosts (P. excavatum) are also described for the first time. The prevalence, abundance and intensity of infection ofH. spinigera in several fish are quan- tified along with the occurrence of P. excavatum, the parasite’s intermediate host, in fish stomach contents. Although H. spinigera’s transmission mode (trophic transmission) and fish diet potentially expose all fish species to infection, some level of host specificity must exist as parasite prevalence, abundance and intensity of infection vary greatly between potential definitive host species. We sug- gest here that the anatomy of the fish digestive tract and especially that of the stomach plays an important role in host suitability for H. spinigera. While P. excavatum is the only intermediate host in Lake Waihola, H. spinigera was found in six different fish species: Aldrichetta forsteri, Galaxias maculatus (Jenyns), Retropinna retropinna, Rhombosolea retiaria Hutton, Perca fluviatilis Linnaeus and Salmo trutta Linnaeus; although typical hedrurid attachment and mating positions were observed only in R. retropinna and A. forsteri. The limited distribution of H. spinigera is most likely due to that of its different host species (intermediate and definitive), all inhabitants of coastal fresh and brackish waters. Keywords: Hedruris spinigera, nematode, host specificity, parasite distribution, amphipod, fish

Hedruris Nitzsch, 1821, the only genus of the family 1989, Bursey and Goldberg 2000, 2007). Currently, the Hedruridae (Nematoda), consists of parasites of the di- development cycles of only three species have been in- gestive tracts of fishes, frogs, salamanders, lizards, and vestigated: H. androphora Nitzsch, 1821, H. ijimai Mor- turtles. Hedruris was first proposed by Nitzsch (1821) ishita, 1926 and H. suttonae Brugni et Viozzi, 2010 (Pet- to include Hedruris androphora from intestine of Tritu- ter 1971, Hasegawa and Otsuru 1979, Brugni and Viozzi rus cristatus (Laurenti) (Caudata, Salamandridae). Until 2010). recently and the description of a new species from Ar- In New Zealand, there have been very few studies of gentina (Brugni and Viozzi 2010), there were 22 docu- larval nematodes, and nematode development cycles are mented nominal species of Hedruris distributed world- seldom described (Moravec et al. 2003). For example, wide (Bursey and Goldberg 2000, 2007). As common Hedruris spinigera Baylis, 1931 is a common freshwater characters, Hedruris species possess four complex lips, parasite typically found in the stomach of fish where the oesophagus���������������������������������������������� not clearly divided into muscular and glandu- larger female attaches itself deeply into the stomach epi- lar portions, and the terminal end of the female is modi- thelium using a recurved hook in its tail (Fig. 1A, Clark fied into an attachment or prehensile organ (Petter 1971, 1978, McDowall 1990). The smaller male lacks the at- Hasegawa and Otsuru 1979, Anderson 2000). Species of tachment structure and moves freely in the stomach lu- Hedruris are distinguished on the basis of the number and men until it finds a female already attached to the host arrangement of caudal papillae in the male, appearance of and curls around her body (Fig. 1B, Clark 1978, Jellyman eggs in females and morphology of cephalic structures, 1989). Taxonomically, Hedruris spinigera is most similar spicules and female caudal hook (Baker 1986, Hasegawa to H. bryttosi Yamaguti, 1935, H. miyakoensis Hasegawa,

Address for correspondence: C. Lagrue, Laboratoire d’Ecologie Fonctionnelle (EcoLab, CNRS UMR 5245), 29 Rue Jeanne Marvig, BP 24349, 31055 Toulouse cedex 4, France. Phone: +33 (0)5 62 26 99 65; Fax: +33 (0)5 62 26 99 99; E-mail: [email protected]

223 lies in the consideration of fish diet and the identification of the intermediate host(s). During recent investigations into the parasites of freshwater fish in southern New Zea- land, adult H. spinigera were observed in the digestive tract of several fish species, and infections were linked to the consumption of a common brackish water crustacean, the amphipod Paracorophium excavatum Thomson. Pre- viously, Luque et al. (2007) described parasitic nematode larvae from P. excavatum from the same location and clas- sified these asHysterothylacium sp.; here, we critically re- examine this identification in the light of morphological and molecular data on adult H. spinigera. The nematode’s sexually mature adults from definitive hosts and imma- ture adults (i.e. larval stages) from intermediate hosts are matched formally for the first time using molecular evidence and described morphologically; further, the life cycle is determined and host specificity of the parasite is discussed from the results on fish diet and H. spinigera prevalence and abundance in the different fish species.

MATERIALS AND METHODS The parasites were found in several samples of the amphipod Paracorophium excavatum and different fish species collected from the same location. Samples were taken between November Fig. 1. Hedruris spinigera Baylis, 1931, light microscopy of 2008 and February 2009 (Austral summer) along a short stretch mature adults from fish definitive hosts. A – large female with of the shore of Lake Waihola (46°01′S, 170°05′E), near Dun- tail end anchored about two-third through the fish host stom- edin, South Island, New Zealand. Lake Waihola is a shallow, ach epithelium, smaller males in the background; B – mature coastal lake situated 10 km from the Pacific Ocean and influ- male and female shown in typical attachment position with male coiled around tail end of female. Specimens were still attached enced twice daily by a fresh to brackish tidal input, depending to fish stomach epithelium when fixed in 70% ethanol. on tides and river flow (Hall and Burns 2002). Only one other amphipod species is present in Lake Waihola, Paracalliope fluviatilis Thomson. However, despite extensive sampling, no 1989 and H. hanleyae Bursey et Goldberg, 2000 with H. spinigera infection was ever found in this species (Lagrue males exhibiting 10 pairs of papillae and females produc- and Poulin 2008a, b). It is also important to note that although ing nonmammillated eggs. Nevertheless, H. spinigera has Hedruris species occur commonly in amphibians (Chandler one pair of precloacal papillae and one pair of adcloa- 1919, Hasegawa and Otsuru 1979, Anderson 2000, Bursey and cal papillae while H. bryttosi lacks precloacal papillae Goldberg 2000), these potential definitive hosts are totally absent (Yamaguti 1935). Similarly to H. spinigera, H. miyakoen- from Lake Waihola. New Zealand amphibian fauna is naturally sis has one (or two) pair of precloacal papillae (Hasegawa very poor and comprises only few species of highly specialized 1989) but the spicules and the anterior end are morpho- and geographically restricted frogs of the genus Leiopelma (Bell 1978, Towns and Daugherty 1994). logically different from those of H. spinigera. Finally, Nine samples of 150 P. excavatum (1350 individuals in total) H. hanleyae has two pairs of precloacal papillae (Bursey were collected at regular intervals during the sampling period. and Goldberg 2000). Amphipods were captured using a push-net (500 µm mesh size) Intriguingly, although H. spinigera is widely distrib- and dredging the surficial sediment whereP. excavatum burrows. uted within New Zealand and can infect a variety of fish, The retained sediment with associated organisms was placed in this parasite occurs only near coastal areas, especially sorting trays from which P. excavatum individuals were collect- in brackish waters (Stokell 1936, Hine 1978, Jellyman ed (Schnabel et al. 2000). Amphipods were kept alive in plastic 1989). No satisfactory explanation has ever been pro- containers filled with local water and brought back to the labora- posed for this geographical restriction although fish diet tory for later dissection. Fish were collected weekly during the and the identity of intermediate host species have been sampling period using gillnets constructed by Oy Lindeman Ab hypothesized to determine the distribution and abundance of Finland and a standard purse seine net. For gillnetting, three 25 m long multi-mesh nets covering the whole water column of H. spinigera in fish hosts. Nevertheless, H. spinigera’s were used. These gillnets were benthic weighted sets with top intermediate host(s) has never been clearly identified al- floats, 1.5 m high and comprised five panels of 13, 25, 38, 56 though several species of crustacean are assumed to be and 70 mm meshes, 5 m long each (Appelberg 2000). Gillnets suitable (Stokell 1936, Griffiths 1976a, Jellyman 1989). being ineffective for catching fish less than 50 mm long, a seine Thus, the key to understanding H. spinigera’s life cy- net was used to capture the smaller fish species and/or juvenile cle and its dynamic of infection in fish definitive hosts specimens that could not be caught with gillnets (Tischler et al.

224 Luque et al.: Host specificity in the nematodeHedruris spinigera

2000). The seine net was 20 m long and 1.5 m high, with a 5 mm fied DNA was purified using an Invitrogen PureLink PCR puri- mesh diameter, and dragged by two people along the lake shore. fication kit and quantified using a Nanodrop spectrophotometer. Again, the net covered the whole water column at all times so no Sequencing was done by the Allan Wilson Centre sequencing fish could escape underneath or above. After capture, fish (see service (Massey University, Palmerston North, New Zealand) Table 1 for details) were killed immediately to stop the digestion using an ABI 3730 DNA Analyzer. Sequences were edited and process and stored on ice until dissection to preserve internal aligned using Sequencher (GeneCodes Corporation) then com- tissues and parasites for future identification. pared to the NCBI database using BLAST to search for genetic Data on H. spinigera occurrence and numbers in amphipod matches (Altschul et al. 1990). and fish hosts were recorded during dissection and used to de- termine the parasite’s prevalence (proportion of infected hosts), mean abundance (mean number of parasites per host) and mean RESULTS intensity (mean number of parasites per infected host) of infec- Morphological analyses and taxonomic description tion in the different host species. Parasitological terminology applied throughout the article follows the recommendations of Hedruris spinigera Baylis, 1931 Bush et al. (1997). Also, to evaluate the exposure of each fish species to this parasite, stomach contents were examined to de- Mature adults (from fish hosts) Fig. 2A–G termine the importance of the intermediate host (P. excavatum) General: Anterior end with 2 lateral pseudolabia and of H. spinigera in the diet of the different fish species. Note that 2 median lips. Lateral pseudolabia each bearing 2 pairs of fishes were identified to species according to McDowall (1990) papillae and one amphid. Median lips irregular. Buccal and fish scientific names follow FishBase (Froese and Pauly 2000). cavity thin-walled. Oesophagus not clearly divided into Individual nematodes from the amphipod host P. excavatum muscular and glandular portions; anterior end with scler- (larval and immature adult stages) and from the fish hosts (adult otized annulus; junction with valved intestine. Cuticle stage of H. spinigera) Retropinna retropinna (Richardson) (Os- thick, with irregular transverse folds and slight transverse meriformes: Retropinnidae) and Aldrichetta forsteri (Valenci- striations. Female with posterior prehensile sclerotized ennes) (Mugiliformes: Mugilidae) were fixed and stored in 70% hook. ethanol for subsequent morphological analyses and taxonomic Male (based on 9 adult specimens): Total length 7500 description. For light microscopy, the specimens were cleared (6580–9170); maximum body width 177 (170–194); width in Amann’s lactophenol (1:1:2:1 phenol: lactic acid: glycerine: at level of excretory pore 137 (125–151). Head 77 (72– water) in which they were kept during measuring and drawing. 86) long. Posterior one-third of body in 2–3 permanent Drawings were made with the aid of a drawing tube attached coils. Oesophagus cylindrical, 1350 (1260–1470) long. to a light microscope. Measurements are given in micrometres (µm) as the mean followed by the size range in parentheses, with Nerve ring 183 (135–221), excretory pore 434 (406–463) the standard deviation or the standard error for large samples. from anterior extremity. Tail 226 (180–254) long, sharply Identification and classification of the specimens to the -ge pointed. Ten pairs of posterior subventral papillae; 1 pair neric level follow Chabaud (1974, 1975), and to the specific precloacal; 1 pair adcloacal (at sides of cloacal opening); level follow Baylis (1931) and Bursey and Goldberg (2000, 8 pairs postcloacal (7 pedunculate subventral and 1 ses- 2007). Voucher specimens (adults from fish definitive hosts and sile lateral). Precloacal subventral surface with rows of immature adults from amphipod intermediate hosts) are depos- rectangular, scale-like bosses extending from near cloaca ited in the Instituto Oswaldo Cruz Helminthological Collection anteriorly. Spicules 200 (181–219), strongly cuticular- (CHIOC), Rio de Janeiro, Brazil and in the collection of the Ota- ized, fused in distal two-fifths. go Museum, Dunedin, New Zealand, all stored in 70% ethanol Female (based on 10 adult specimens): Total length (Luque et al. 2007). 10120 (7960–12280); maximum body width 404 (346– Eight individual nematodes were stored separately in 95% ethanol for subsequent DNA extraction and genotyping: 4 adult 434); narrow anteriorly, head 89 (85–94) long; width at specimens, one from each of four R. retropinna stomachs and excretory pore 212 (181–227), posterior end swollen. Oe- 4 larval specimens from four different amphipods (P. excava- sophagus 1900 (1820–2070) long. Nerve ring 241 (208– tum) were isolated in Eppendorf tubes. DNA was extracted from 270), excretory pore 491 (386–559) from anterior extrem- the nematodes using 400 µl of 5% Chelex solution (BioRad) ity. Vulva 1220 (1120–1290), anus 388 (351–415) from containing 0.1mg/ml Proteinase K (Roche), incubated at 60 °C posterior end. Prodelphic uterus leads to a single ovijector overnight, followed by 10 min at 90 °C. Approximately 900 bp that joins common trunk of uterus. Ovaries thin, taper- of 18S and 400 bp of the third domain of 28S (D3) were ampli- ing tubes that coil about anterior portion of digestive tract fied using the primers Nem_18S_F and Nem_18S_R (Floyd et but not reaching oesophageal-intestinal junction level. al. 2005) and D3a and D3b (Sharpe 1999), respectively, on each In gravid specimens, egg-filled uteri occupy entire body of the eight specimens. Each 25 µl PCR contained 0.5 µM of cavity. Tail curved dorsally, formed into a sucker-like ap- each primer, 0.8 mM dNTPs, 1.5 mM MgCl2, 1 unit BioTaq Red (Bioline) and 1 µl of extracted DNA. PCR amplifications were paratus armed with large sclerotized hook 112 (105–121) performed using an Eppendorf Mastercycler ep gradient S un- long. Eggs 28.5 ± 2.4 (22.5–33.3) long, 13.1 ± 1.8 (11– der the same conditions for both genes, consisting of 94 °C for 16) wide (mean ± S.D.; n = 30), without lateral “mammil- 3 min, followed by 45 cycles of 94 °C for 30 s, 45 °C for 30 s and lae”, cylindrical, operculated at the poles, containing fully 72 °C for 60 s, with a final extension of 72 °C for 240 s. Ampli- developed larva.

225 Fig. 2. Hedruris spinigera Baylis, 1931, light microscopy of mature adults from fish definitive hosts.A – male, anterior end of body, ventral view; B – female, anterior end of body, lateral view; C – male, posterior end, lateral view; D – male, spicule, lateral view; E – female, posterior end, lateral view; F – female, vulva (indicated by arrow) and terminal portion of vagina, eggs, lateral view; G – female, tail hook, lateral view.

226 Luque et al.: Host specificity in the nematodeHedruris spinigera

H o s t : Retropinna retropinna (Richardson) (Osmeriformes: (Instituto Oswaldo Cruz Helminthological Collection, Bra- Retropinnidae) and Aldrichetta forsteri (Valenciennes) zil, CHIOC No. 35671). (Mugiliformes: Mugilidae) were the only hosts containing females of the parasite attached to host gastric tissues and Genetic analyses males coiled around the females in mating position. How- ever, four other species contained unattached live worms; Up to 755 bp of 18S and 317 bases of 28S were se- Galaxias maculatus (Jenyns), Rhombosolea retiaria Hutton, quenced for each individual and the 8 sequences (4 sexu- Perca fluviatilis Linnaeus and Salmo trutta Linnaeus. ally mature adults and 4 immature adults of H. spinig- S i t e o f i n f e c t i o n : Stomach is the only site of attach- era) were submitted to the GenBank database (GenBank ment for female specimens, but the parasite can also be found Acc. Nos. HM484336 to HM484351). All sequences were free-living (i.e. unattached) in the intestine and rectum of fish identical and a BLAST search confirmed the correct gene hosts. regions had been amplified. Consequently, all specimens L o c a l i t y : Lake Waihola (46°01′S, 170°05′E), South Island, (4 adults and 4 immature adults) can be considered as be- New Zealand. ing of the same species. There were no identical matches P r e v a l e n c e a n d I n t e n s i t y : Depending on the host in NCBI for either sequence. For 18S, the closest matches species (see Table 2). were to species in the order Spirurida (90–92% identity); D e p o s i t i o n o f v o u c h e r s p e c i m e n s : Six adult Turgida torresi (Spirurida: Physalopteridea, GenBank male and six adult female specimens (Instituto Oswaldo Cruz Acc. No. EF180069) and Physaloptera turgida (Spiru- Helminthological Collection, Brazil, CHIOC No. 35669 and rida: Physalopteridea, GenBank Acc. No. DQ503459). 35670 for males and females respectively). Spirurid nematodes are numerous and widespread ob- ligatory gastrointestinal parasites of vertebrates (Stunkard Immature adults (in amphipod host) Fig. 3A–D 1953, Chabaud and Bain 1994). The 28S (D3) sequence General: Morphological characteristics of immature showed much weaker matches and also lower coverage. adults similar to mature adults from fishes, except by their The top two matches were also to spirurid nematodes; smaller size and absence of eggs. Cystidicola farionis (Cystidicolidae, GenBank Acc. No. Male (based on 3 adult immature specimens): Total AY161299) and Cystidicola stigmatura (Cystidicolidae, length 6670 (5580–7770); maximum body width 171 GenBank Acc. No. AY161298). (165–174); width at level of excretory pore 133 (129–135). Head 57 (54–63) long. Oesophagus 1340 (1280–1410) Hedruris spinigera prevalence, abundance and long. Nerve ring 255 (n = 1), excretory pore 405 (359– intensity of infection in fish hosts 437) from anterior extremity. Tail 303 (277–347) long. Fish from 7 different species were captured in Lake Spicules 185 (119–230) long. Waihola during the sampling period: Aldrichetta forsteri Female (based on 3 adult immature specimens): To- (Valenciennes), Galaxias maculatus (Jenyns), Gobiomor- tal length 7290 (5640–8340); maximum body width phus cotidianus McDowall, Retropinna retropinna (Rich- 303 (301–306); head 74 (68–77) long; width at excretory ardson), Rhombosolea retiaria Hutton, Perca fluviatilis pore level 169 (117–171). Oesophagus 1750 (1430–2120) Linnaeus and Salmo trutta Linnaeus (see Table 1 for de- long. Nerve ring 275 (216–334), excretory pore 509 tails). Perca fluviatilis and S. trutta were introduced from (387–686) from anterior extremity. Vulva 906 (815–984), Europe and are now widespread and abundant in New anus 415 (396–443) from posterior end. Sclerotized cau- Zealand (Thompson 1922). dal hook 129 (124–133) in length. Gobiomorphus cotidianus was the only fish species Intermediate host: Paracorophium excavatum Thom- free of H. spinigera (Table 2). At least some individuals son (: ) is the only intermediate host of the remaining species contained the nematode in their species recognized at the study site. digestive tract, although infection parameters of H. spini- S i t e o f i n f e c t i o n : Haemocoelomic cavity. Worms coiled gera varied greatly between host species (Table 2). The or folded tightly but not encysted. prevalence (93.1 and 46.9%), abundance (mean ± S.E.; L o c a l i t y : Lake Waihola (46°01′S, 170°05′E), South Island, 29.2 ± 8.7 and 12.1 ± 7.5) and intensity (mean ± S.E.; New Zealand. 31.4 ± 8.9 and 25.9 ± 10.6) of H. spinigera were clearly P r e v a l e n c e a n d i n t e n s i t y : Between 2.7 and 14.7% higher in A. forsteri and R. retropinna, respectively, than (mean ± S.E. = 7.3 ± 1.3) of hosts were infected, depending in any other fish species (Table 2). Although H. spinig- on the sample (n = 150 P. excavatum per sample), usually era’s prevalence varied between 5.9 and 39.1% in the oth- with 1 worm (mean ± S.E. = 1.11 ± 0.03), but sometimes 2 and exceptionally 3 (one occurrence) H. spinigera indi- er fish host species, the numbers of nematode individuals viduals were found. per host were clearly lower than in A. forsteri and R. retro- D e p o s i t i o n o f v o u c h e r s p e c i m e n s : Three im- pinna (Table 2). More importantly, A. forsteri and R. ret- mature adult males and 6 larval specimens (Otago Mu- ropinna were the only fish species in which H. spinigera seum, New Zealand, collection No. IV16246 to IV16254). attained full size and maturity, and where female H. spini- Two immature adult females and two immature adult males gera established and attached to the stomach epithelium

227 Fig. 3. Hedruris spinigera Baylis, 1931, immature adults from amphipod intermediate hosts. A – female, anterior end, lateral view; B – female, posterior end, lateral view; C – male, posterior end, lateral view; D – male, spicule, lateral view. with males coiled around them in mating position. In no DISCUSSION other fish species, potential definitive hosts, were female Adult specimens of Hedruris spinigera were adequate- parasites found attached or males associated with females ly described by Baylis (1931). Here, the mature adults of in the typical mating position. H. spinigera display morphological characteristics con- While P. excavatum was found in the diet of all fish sistent with the original species description although adult species, occurrence and abundance of the amphipod in- males described in this study have smaller spicules, and termediate host in fish stomach contents are also highly adult females have smaller tail hook and eggs, when com- variable (Table 2). Interestingly, A. forsteri and R. retro- pared to the specimens described previously from Lake pinna, which are the most heavily H. spinigera-infected Ellesmere (43°47′S, 172°23′E), a brackish costal lake species, were not the main consumers of P. excavatum, the near Christchurch, South Island, New Zealand (Baylis intermediate host of H. spinigera (Table 2). Also, while 1931). Unlike other helminths, H. spinigera preferentially the prevalence of H. spinigera was 0% in G. cotidianus, attaches to the stomach epithelium of fish definitive hosts 17.1% of these fish contained P. excavatum in their stom- and more specifically in the pyloric region (Hine 1980). ach and could have been exposed to the parasite. The bot- Therefore, individuals observed in the intestine and rec- tom-dwelling R. retiaria feeds heavily on P. excavatum, tum of fish are likely to be rapidly passed out. Consequent- as indicated by the results in Table 2, but does not seem to ly, lower H. spinigera prevalence, mean abundance and/ be a very suitable host for H. spinigera.

228 Luque et al.: Host specificity in the nematodeHedruris spinigera

Table 1. Fish species sampled in Lake Waihola with the sample size (n = number of fish dissected), and body length (mean ± S.E., with range in parentheses) given for each fish species.

Fish species Order Family n Body length (mm) Aldrichetta forsteri Mugiliformes Mugilidae 29 185.5 ± 6.6 (138–260) Galaxias maculatus Osmeriformes Galaxiidae 34 69.4 ± 2.3 (45–105) Gobiomorphus cotidianus Perciformes Eleotridae 35 44.4 ± 1.5 (32–72) Perca fluviatilis Perciformes Percidae 68 132.2 ± 10.1 (36–360) Retropinna retropinna Osmeriformes Retropinnidae 32 80.4 ± 1.9 (65–104) Rhombosolea retiaria Pleuronectiformes Pleuronectidae 46 90.5 ± 14.3 (30–356) Salmo trutta Salmoniformes Salmonidae 23 397.8 ± 21.5 (215–690)

Table 2. Hedruris spinigera prevalence (proportion of hosts containing the parasite), mean abundance (mean number of parasites per host examined ± S.E.) and mean intensity (mean number of parasites per host infected ± S.E.), and Paracorophium excavatum occurrence (proportion of fish containing the prey in their diet), mean abundance (mean number of prey per fish examined ± S.E.) and mean intensity (mean number of prey per individual fish containing the prey in its diet ± S.E.) in the different fish species captured in Lake Waihola.

Hedruris spinigera Paracorophium excavatum Fish species Prevalence (%) Abundance Mean intensity Occurrence (%) Abundance Mean intensity A. forsteri 93.1 29.2 ± 8.7 31.4 ± 8.9 34.5 3.4 ± 1.2 9.8 ± 2.5 G. maculatus 5.9 0.1 ± 0.1 2.0 ± 1.0 2.9 0.03 ± 0.03 1.0 ± n. a. G. cotidianus 0.0 n. a. n. a. 17.1 1.1 ± 0.6 6.7 ± 3.0 P. fluviatilis 14.7 0.2 ± 0.0 1.0 ± n. a. 35.3 4.5 ± 1.3 12.7 ± 3.0 R. retropinna 46.9 12.1 ± 7.5 25.9 ± 10.6 15.6 2.2 ± 1.1 13.8 ± 4.8 R. retiaria 26.1 4.6 ± 2.2 17.7 ± 7.2 78.3 20.0 ± 9.4 25.5 ± 11.8 S. trutta 39.1 3.1 ± 1.4 8.0 ± 2.9 26.1 16.2 ± 13.0 62.0 ± 47.9 or mean intensity in some fish species may be due to their fail to attach in the new hosts and are quickly evacuated. digestive tract not providing appropriate attachment sites Generally, some fish species in which H. spinigera is to the parasite (Campos and Carbonell 1994); although found are clearly not very suitable for the adult para- other factors like gastrointestinal pH, digestive enzymes site (McDowall 1990). As suggested previously (Stokell and even specific parasite behaviour can affect parasite at- 1936, Griffiths 1976a), and as indicated by our results tachment and establishment in hosts (Rogers 1960, Irwin (prevalence, abundance and intensity of H. spinigera in- 1997). For example, G. cotidianus and R. retiaria, in con- fection), A. forsteri and R. retropinna seem to be the only trast with other species like A. forsteri and R. retropinna truly suitable hosts for this parasite. which possess a well defined pylorus (Hine 1978, Horn Interestingly, A. forsteri and R. retropinna are typically 1989), display extremely simple alimentary canals with coastal species. Aldrichetta forsteri is mainly a marine a poorly defined stomach that might not be adequate for fish but very commonly penetrates several kilometres in- H. spinigera establishment. In that sense, it is worth not- land into fresh or brackish water where it becomes abun- ing that, although this species was previously classified dant during the summer months and feeds on the available as a suitable definitive host for this parasite (Hine 1980, aquatic invertebrates (Eldon and Greager 1983, McDow- Jellyman 1989), H. spinigera was never recovered from all 1990). Retropinna retropinna populations are often G. cotidianus in Lake Waihola despite extensive sampling anadromous, migrating from the sea to spawn in coastal (Lagrue and Poulin 2008a, present study). Similarly, the freshwater systems during spring and early summer (Mc- U-shaped structure and very thick epithelium of the P. flu- Dowall 1990). In Lake Waihola, R. retropinna are most viatilis and S. trutta stomachs might not be suitable for likely a mixture of anadromous and resident (non-migra- H. spinigera attachment (Griffiths 1976b). tory) stocks as fish densities vary greatly between sum- Although most fish species consumed H. spinigera’s mer and winter but adult fish are present all year round. intermediate host and this parasite seems to be a general- Retropinna retropinna feed heavily on available benthic ist with respect to its fish hosts, it occurs either at low or pelagic invertebrate prey (Eldon and Greager 1983, abundance or only temporarily in several fish species. For McDowall 1990). The geographical distribution of these example, the presence of H. spinigera in predatory fish two fish species, H. spinigera’s definitive hosts, might like Anguilla australis Richardson, Anguilla dieffenbachii therefore explain why the distribution of the parasite is Gray, S. trutta and P. fluviatilis is considered to be a sec- limited to coastal areas (Stokell 1936, Hine 1978, Jelly- ondary transfer from consumed prey fish, such as heavily man 1989). infected individuals of R. retropinna (Stokell 1936, Grif- For the first time, larval stages (immature adults) and fiths 1976a, Hine 1978). However, these parasites mostly intermediate host species of H. spinigera are formally

229 identified and matched. Genetic analyses confirmed that phologically similar to mature adults (Petter 1971, Brugni larval nematodes found in amphipod hosts are the same and Viozzi 2010). This similitude was also observed in species as adult H. spinigera described from fish defini- H. spinigera between immature adults found in amphipod tive hosts. Although previous studies classified the small- intermediate hosts and adults from fish definitive hosts. er amphipod Paracalliope fluviatilis as the most likely In the case of H. spinigera, precocious maturity and rapid intermediate host for H. spinigera (Jellyman 1989, Hine gamete production after transmission could have evolved et al. 2000), extensive sampling of this potential host with the use of definitive hosts that have seasonally vari- species in Lake Waihola failed to find H. spinigera in- able densities. fections despite very high P. fluviatilis densities (Lagrue Here, we describe larval stages (immature adults) of and Poulin 2008a, b). Here, the larger benthic amphipod Hedruris spinigera, identify the intermediate host species Paracorophium excavatum was the only intermediate involved in its life cycle, and present data on prevalence, host identified for H. spinigera. The parasite must reach mean abundance and mean intensity of H. spinigera in- its definitive host through trophic transmission; i.e. when fection in the different species of fish, potential definitive H. spinigera-infected P. excavatum are consumed by fish hosts. The results seem to indicate that both intermediate definitive hosts. Again, the distribution of H. spinigera, and definitive host distributions may limit the geographi- limited to coastal areas, may be partly attributed to the cal distribution of the parasite. Indeed, hosts and parasites geographical distribution of the intermediate host P. ex- are found in the same type of habitats (coastal fresh and cavatum, a common crustacean species from coastal la- brackish water systems) and likely have geographic rang- goons and brackish habitats around New Zealand (Chap- es broadly overlapping, although geographically exhaus- man 2002, Luque et al. 2007). tive sampling is needed to verify similarities between host This study is also the first record of immature adults of and parasite distributions. We also suggest that H. spini- H. spinigera in the intermediate host after Baylis (1931) gera infection patterns in fish hosts could be due to host seriously questioned whether such precocious sexual de- suitability rather than the probability of transmission velopment existed. Luque et al. (2007) described parasitic given that all fish species consume the intermediate host nematode larvae from P. excavatum from the same loca- Paracorophium excavatum and are therefore exposed to tion and classified these as Hysterothylacium sp. How- the parasite. As mentioned herein, the anatomy of fish di- ever, examination of the present specimens and of fig. gestive tracts, especially that of the stomach, is likely to 2A–C included in Luque et al. (2007) showed clearly that play an important role in host suitability for H. spinigera the larval nematodes collected from P. excavatum were in although further investigations are needed to test for this fact immature adults of H. spinigera identical to the speci- hypothesis. mens described herein; although it is possible that larval Hysterothylacium sp. also occur in amphipods of Lake Acknowledgements. We thank Antoine Lecerf (EcoLab CNRS Waihola. Both Hedruris androphora and H. ijimai were UMR 5245) for comments on an earlier version of the manu- previously shown to develop into immature adults within script and Ken Miller (Drawing Office, Department of Zoology, isopod intermediate hosts (Anderson 2000). Similarly to University of Otago) for help on the artwork. This research was H. spinigera, larvae of the newly described species He- supported by a PBRF Research Enhancement grant from the Zoology Department, University of Otago, to Robert Poulin. druris suttonae also exhibited extreme precocity in their Fabiano M. Vieira was supported by a Doctoral fellowship from amphipod intermediate host (Brugni and Viozzi 2010). REUNI/UFRRJ (Programa de Apoio à Reestruturação e Expan- This was suggested to accelerate gamete production in the são das Universidades Federais, Brasil) and Kristin Herrmann definitive host (Petter 1971, Hasegawa and Otsuru 1979, by a University of Otago PhD scholarship. José L. Luque was Anderson 2000). Only H. androphora and H. suttonae im- supported by a Research fellowship from CNPq (Conselho Na- mature adults were previously described and found mor- cional de Pesquisa e Desenvolvimento Tecnológico, Brasil).

REFERENCES

Altschul S.F., Gish W., Miller W., Myers E.W., Lipman D.J. Baylis H.A. 1931: A species of the nematode genus Hedruris occur- 1990: Basic local alignment search tool. J. Mol. Biol. 215: 403– ring in the trout in New Zealand. Ann. Mag. Nat. Hist., Ser. 10, 410. 7: 105–114. Anderson R.C. 2000: Nematode Parasites of Vertebrates. Their De- Bell B.D. 1978: Observations on the ecology and reproduction of the velopment and Transmission. 2nd Ed. CABI Publishing, Walling- New Zealand leiopelmid frogs. Herpetologica 34: 340–354. ford, UK, 672 pp. Brugni N.L., Viozzi G.P. 2010: A new hedrurid species (Nematoda) Appelberg M. 2000: Swedish standard methods for sampling fresh- from galaxiid fishes in Patagonia (Argentina), and infection of am- water fish with multi-mesh gillnets. Natl. Board Fish. 1: 1–27. phipods as intermediate host. J. Parasitol. 96: 109–115. Baker M.R. 1986: Revision of Hedruris Nitzsch (Nematoda: Bursey C.R., Goldberg S.R. 2000: Hedruris hanleyae n. sp. (Nem- Habronematoidea) from aquatic vertebrates of North America. atoda: Hedruridae) from Hemidactylus garnotii (Sauria: Gekko- Can. J. Zool. 64: 1567–1572. nidae) from the Cook Islands, Oceania. J. Parasitol. 86: 556–559.

230 Luque et al.: Host specificity in the nematodeHedruris spinigera

Bursey C.R., Goldberg S.R. 2007: New species of Hedruris (Nem- Hine P.M., Jones J.B., Diggles B.K. 2000: A checklist of parasites atoda: Hedruridae), Anuracanthorhynchus lutzi (Hamann, 1891) of New Zealand fishes, including previously unpublished records. n. comb. and other helminths in Lithobates warszewitschii (Anu- NIWA Technical Report 75, 93 pp. ra: Ranidae) from Costa Rica. Carib. J. Sci. 43: 1–10. Horn M.H. 1989: Biology of marine herbivorous fishes. Oceanogr. Bush A.O., Lafferty K.D., Lotz J.M., Shostak A.W. 1997: Para- Mar. Biol. Annu. Rev. 27: 167–272. sitology meets ecology on its own terms: Margolis et al. revisited. Irwin S.W.B. 1997: Excystation and cultivation of trematodes. In: J. Parasitol. 83: 575–583. B. Fried and T.K. Graczyk (Eds.), Advances in Trematode Biol- Campos A., Carbonell E. 1994: Parasite community diversity in ogy. CRC Press, Boca Raton, Florida, pp. 57–85. two Mediterranean labrid fishes Symphodus tinca and Labrus Jellyman D.J. 1989: Occurrence of the nematode Hedruris spinigera merula. J. Fish Biol. 44: 409–413. in the stomachs of freshwater eels. N. Z. J. Zool. 16: 185–189. Chabaud A.G. 1974: Class Nematoda. Keys to subclasses, orders and Lagrue C., Poulin R. 2008a: Lack of seasonal variation in the life- superfamilies. In: R.C. Anderson, A.G. Chabaud and S. Willmott history strategies of the trematode Coitocaecum parvum: no ap- (Eds.), CIH Keys to the Nematode Parasites of Vertebrates. Com- parent environmental effect. Parasitology 135: 1–9. monwealth Agricultural Bureau, Farnham Royal, United King- Lagrue C., Poulin R. 2008b: Intra- and interspecific competition dom, pp. 6–17. among helminth parasites: effects on Coitocaecum parvum life- Chabaud A.G. 1975: Keys to the genera of the order Spirurida. history strategy, size and fecundity. Int. J. Parasitol. 38: 1435– Spiruroidea, Habronematoidea and Acuarioidea. In: R.C. Ander- 1444. son, A.G. Chabaud and S. Willmott (Eds.), CIH Keys to the Nema- Luque J.L., Bannock L.M., Lagrue C., Poulin R. 2007: Larval tode Parasites of Vertebrates, Part 2. Commonwealth Agricultural Hysterothylacium sp. (Nematoda, Anisakidae) and trematode Bureau, Farnham Royal, United Kingdom, pp. 29–58. metacercariae from the amphipod Paracorophium excavatum Chabaud A.G., Bain O. 1994: The evolutionary expansion of the (Corophiidae) in New Zealand. Acta Parasitol. 52: 146–150. Spirurida. Int. J. Parasitol. 24: 1179–1201. McDowall R.M. 1990: New Zealand Freshwater Fishes: a Natural Chandler A.C. 1919: On a species of Hedruris occurring com- History and Guide. Heinemann Reed MAF Publishing Group, monly in the western newt, Notophthalmus torosus. J. Parasitol. Auckland, New Zealand, 230 pp. 5: 116–122. Moravec F., Fredensborg B.L., Latham A.D.M., Poulin R. 2003: Chapman M.A. 2002: Australasian species of Paracorophium Larval Spirurida (Nematoda) from the crab Macrophthalmus hir- (Crustacea: Amphipoda): the separate identities of P. excavatum tipes in New Zealand. Folia Parasitol. 50: 109–114. (Thomson, 1884) and P. brisbanensis sp. nov. J. R. Soc. N. Z. 32: Nitzsch C.L. 1821: Ascaris. In: J.S. Ersch and J.G. Gruber (Eds.), 203–228. Allgemeine Enzyklopedie der Wissenschaften und Künste, Leip- Clark W.C. 1978: Head morphology and prehensile adaptations in zig, pp. 6.44–6.49. Hedruris spinigera (Nematoda: Spirurida). N. Z. J. Zool. 5: 497– Petter A.-J. 1971: Redescription d’Hedruris androphora Nitzsch, 501. 1821 (Nematoda, Hedruridae) et étude de son développement chez Eldon G.A., Greager A.J. 1983: Fishes of the Rakaia lagoon. N. Z. l’hôte intermédiaire. Ann. Parasitol. Hum. Comp. 46: 479–495. Ministr. Agric. Fish., Fish. Environ. Rep. 30: 1–65. Rogers W.P. 1960: The physiology of infective processes of nema- Floyd R.M., Rogers A.D., Lambshead P.J.D., Smith C.R. 2005: tode parasite; the stimulus from the animal host. Proc. R. Soc. Nematode-specific PCR primers for the 18S small subunit rRNA Lond., Ser. B, 152: 367–386. gene. Mol. Ecol. Notes 5: 611–612. Schnabel K.E., Hogg I.D., Chapman M.A. 2000: Population ge- Froese R., Pauly D. 2000: FishBase 2000: Concepts, Design and netic structure of two New Zealand corophiid amphipods and Data Sources. ICLARM, Los Baños, Laguna, Philippines, 344 pp. the presence of morphologically cryptic species: implications for Griffiths W.E. 1976a: Hedruris spinigera (Nematoda: Spiruridae) the conservation of diversity. N. Z. J. Mar. Freshwater Res. 34: in stomachs of perch Perca fluviatilis L. N. Z. J. Mar. Freshwater 637–644. Res. 10: 533–536. Sharpe R.G., Hims M.M., Harbach R.E., Butlin R.K. 1999: PCR- Griffiths W.E. 1976b: Food and feeding habits of European perch in based methods for identification of species of the Anopheles min- the Selwyn River, Canterbury, New Zealand. N. Z. J. Mar. Fresh- imus group: allele-specific amplification and single-strand confor- water Res. 10: 417–428. mation polymorphism. Med. Vet. Entomol. 13: 265–273. Hall C.J., Burns C.W. 2002: Environmental gradients and zoo- Stokell G. 1936: The nematode parasites of Lake Ellesmere trout. plankton distribution in a shallow, tidal lake. Arch. Hydrobiol. Trans. Proc. R. Soc. N. Z. 66: 80–96. 154: 485–497. Stunkard H.W. 1953: Life histories and systematics of parasitic Hasegawa H. 1989: Neoentomelas asatoi gen. and sp. n. (Nematoda: worms. Syst. Zool. 2: 7–18. Rhabdiasidae) and Hedruris miyakoensis sp. n. (Nematoda: He- Thomson G.M. 1922: The Naturalization of and Plants in druridae) from skinks of the Ryukyu Archipelago, Japan. Proc. New Zealand. Cambridge University Press, Cambridge, UK, 607 Helminthol. Soc. Wash. 56: 145–150. pp. Hasegawa H., Otsuru M. 1979: Life history of an amphibian nema- Tischler G., Gassner H., Wanzenbock J. 2002: Sampling charac- tode, Hedruris ijimai Morishita, 1926 (Hedruridae). Jpn. J. Para- teristics of two methods for capturing age-0 fish in pelagic lake sitol. 28: 89–97. habitats. J. Fish Biol. 57: 1474–1487. Hine P.M. 1978: Distribution of some parasites of freshwater eels in Towns D.R., Daugherty C.H. 1994: Patterns of range contractions New Zealand. N. Z. J. Mar. Freshwater Res. 12: 179–187. and extinctions in the New Zealand herpetofauna following hu- Hine P.M. 1980: Distribution of helminthes in the digestive tracts of man colonisation. N. Z. J. Zool. 21: 325–339. New Zealand freshwater eels. 1. Distribution of digeneans. N. Z. Yamaguti S. 1935: Studies on the helminth fauna of Japan. Part 9. J. Mar. Freshwater Res. 14: 329–338. Nematodes of fishes, 1. Jpn. J. Zool. 6: 338–386.

Received 27 January 2010 Accepted 26 March 2010

231